There is growing need for information of embodied energy impacts. As I argued in a Linkedin article last week: Needed: A international Embodied Energy Data Authority’.*
But here I want to address the lack of data for especially installations. Mainly due to the need to develop energyneutral or energy positive new houses, as well as mass retrofitting existing houses. Which requires supposedly to invest in thick insulation layers, triple glass , combined with for instance low temperature conversion systems, and/or infra-red systems, heat pumps etc. Pushing both embodied energy and materials investments to a much higher level. For so called zero energy houses, ZEB or net-ZEB, energy is in fact not really an issue any more, regarding heating and ventilation, since based on renewable energy (averaged over a year), the main decisive choices are by the invested materials impact.[1,2] Its all about searching for the optimum, lowest impact , in a combination of reduction measures and production /conversion measures.
A interesting illustration of this optimisation process was developed in the EU- More Connect project. A existing house was compared for 4 different ways of creating a 0-energy house. From no measures ( and only suplying renewable energy), via basic reduction measures and a 5 cm outside insulation layer to a passive style retrofit (with each fewer solar panels). Its showed that all these strategies were better then no measures, but that optimal strategy was basic energy reduction measures with medium amount of solar panels. Of course the passive style retrofit is better then no retrofit. But its not the optimum! Its always nescesarry to evaluate several strategies to find the optimum, which is, in case of a ZEB retrofit, only determined by the input of embodied energy, for reduction measures and RE production measures combined. (see a PPT presentation and xls-file here )
A alternative approach is to reduce heated area, to limit the material investments, ie, not retrofitting the whole house, but a section. This requires some life style adaptation.  But would make much more sense, since the standard retrofit measures are laid out for the worst period in the climate year, while for instance in the Netherlands, there are are only a few days a year with freezing temperatures ( and decreasing). Its not smart to retrofit the whole house as if it freezes the whole year.
Anyway, to make good choices and strategies data are important, especially for practice: more and more practitioners are aware of this problem, and are searching for sensible solutions. However, for instance for installations , heatpumps etc, good data are lacking. Asking colleagues for information on data for installations, I collected some material, which I will summarize below. Its not a scientific analyses, it merely tries to give some insight of available data, reference projects and literature. Its some first collection that could evolve into some practical guidance for daily use .
See at the end reference datasheet, based on own calculations, experiences with projects and data from for instance annex 57 work, which I use for students to give them something to hold on, while making design decisions. I welcome better or additional data, and references.
A footnote with all this: Most data I received or retrieved, are in GWP-Global Warming Potential / embodied carbon, that is in gr CO2-eq. In fact a book has just been published on this issue by some valued colleagues in the field: Embodied Carbon in Buildings . At the moment of writing I have not yet read it myself, but a review is available.  .
However, I prefer embodied energy data. For several reasons. First of all CO2 is a end of pipe effect, a consequence, rather then a cause. If we focus on 1 effect, we will create other problems: Its better to optimise on energy, and tackle more problems at once.
Secondly, even if we create zero carbon operational energy houses, there is still a lot of impact by the materials investments. And even if the materials are zero carbon produced, the renewable energy devices for that require a lot of material and again energy to be produced: The less the better.
Lastly, suppose globally we only use renewable energy, then still its wise to optimise for ( low) energy, since all energy requires materials investment , which require ( renewable) energy to produce etc. Its energy , and not carbon that determines the impact, especially in a renewable energy world . ( And limiting energy also reduces Carbon impact , as long as we still produce some with fossils)
So I am looking especially for embodied energy data, not embodied carbon. Besides, In fact the best solution would be, in case of building/construction/installation products, to have energy end use data, not primary energy, so that it is easier to compare different energy management scenarios at building level, regardless a national energy supply system, and make direct building data easier comparable between countries for instance. It also makes it easier to, for instance, involve labor again in the decision making process.
Note: Its important to notice that the usual calcuation of embodied energy and or GWP in CO2-eq, is done by a whats called ‘ process analyses’, counting only direct upstream inputs. However, in a ‘input output’ analyses all secondary or tertiary inputs count as well, and Embodied energy could rise by factors! . ( By the way for operational energy its the same, usually only direct input is calculated). So the situation is much worse as usually depicted, But here we first try to get the data right on process level for installations.
Beware: an important parameter in embodied energy or embodied carbon data for buildings is how the functional unit is defined: how the m2’s are calculated. In principle the total embodied energy impact over all construction work should be divided , especially for heating and ventilation, by only the heated m2, and even better , only by net rentable and heated area . ( that is: including verandas, balconies, garages, or even parking garages in the embodied energy count, but not in the m2 counting) In the examples below its not always clear how m2 is calculated.
To be further developed. RR November 2020
Deutsche Umwelt Bundesambt pilot-Berlin office
A first indication about the impact form Installations comes from a Demonstration office for the Deutsche Umwelt Bundesambt, well documented, in Berlin. Incidentally also a zero energy office with regard to building-related energy. 
The figures show that the embodied energy from total installations accounts for no less than 25% of the total embodied energy, and even 40% of the CO2 emissions. That is quite substantial, in absolute numbers the EE is 2.7 GJ / m2 floor (for the installations, incl PV panels). The building as a whole comes out at 10 GJ / m2 of useful floor space.  That is well above average, most common buildings are around 5-6 GJ / m2 floor, (excluding installations).
A second example is the Swiss Siemens head office. 7 floors, incl 2 underground parking areas. Building 32000 m2, of which 22000 heated. Reference life time is 60 years. Detailed material quantities for the installations are available, see table for the HVAC systems.
“The total embodied impact for the HVAC systems of this case study is 183 kgCO2eq/m2. Compared to the existing knowledge for the total embodied impact of office buildings, the HVAC embodied impact would be in the range of 15–36%, which is significantly higher than previous studies and estimations”.
Since I am interested in the embodied energy, I made some own calculations based on the material data. The table adds up, to the following figures: total mass for HVAC: 409.106 kg, that is: 22.28 kg/m2 heated floor.
This gives a total Embodied energy of: 26812318 MJ, that is 1,22 GJ/m2 . This is however only based on the materials, not on manufacture and transport of final products. The real figure will be higher. Its also exclusive of Solar panels and bos, in case the building would be fitted for net zero operation. (Should be added when compairing to the Berlin office).
Solar panels Embodied Energy varies from 2,2-6 GJ/m2 for panels and BOS. (bos is 600-900 MJ/m2, depending roof or ground mounting) (bos= balance of system equipment, =(local) cables, inverters etc.). See this paper for more detailed data. 
A somewhat easy accessible database with figures on installations , is data from the German Oekobau database:  Mind that some are based on EPD’ Environmental product declaration submitted by product manufacturers .
For instance: a 10 kW heatpump, air to water, electric: weight: 134 kg, and the embodied energy is given as EEpe =6977 MJ /unit (est lifetime: 20 years est. COP 3,1) .
For a flat solar collector of 1m2 the following data are available: weight: 18,27 kg/m2, and the embodied energy is (A1-A3) EEpe =1819.9 MJ/m2 .(est. liftimet:20 y)
For a decentralized Ventilation unit with heat recovery (wall & ceiling mounted) at 60 m3/h , the weight is 3,69 kg, and the embodied energy: EE= 481 MJ/piece.
detailed scientific analyses
There are a more scientific data and research on the issue of building installations published, however mainly in GWP and CO2-eq. For instance by Frishknecht and Kiamili. For the diehard calculators among us, there are lots of details in the publications, to make separate calculations. See literature. [13-x]
Generic data by CLF
The so called Carbon Leadership Forum provides some data on comparing buildings for mechanical, electrical and plumbing (MEP) input. ( in GWP-CO2-eq. ) Mechanical is by far the highest impact of the three, up to 75 % (TI=finishes,furniture and fixtures) More details in docs: http://www.carbonleadershipforum.org/projects/lca-of-mep-and-ti/
Houses EE data inventory
The resulting embodied energy use in Dutch dwelling archetypes varies from 52 to 106 MJ/(m²·a), annualised over building lifetimes, which implies 3.0 to 6.4 GJ/m2 in total. Mind that energy invested and related emissions are not spread over the years, they are direct at the start. These values are for the building construction and exclude recurrent embodied energy, solar panels and technical installations. For operational energy use the range is 124 to 682 MJ/(m2·a). ‘A total energy use reduction of 36% can be reached in 2050 through 46% reduction in operational energy use and 35% increase in embodied energy use, compared to 2015′. This research confirms that the relative importance of embodied energy use is increasing fast, absolute and relative: the embodied energy use in standard homes is about 10–12% of the total energy use, while it is 36–46% in energy efficient homes.  In some cases its already over 50 %, and in any case, if we take 2050 as the moment of net zero emissions, Embodied energy is already higher as operational energy most of the time.
A recent international meta study on published scientific papers, shows for new houses an average for european houses , recalculated form GWP data, of around 4,1 GJ /m2 net heated area, excl. installations, and excl. recurrent embodied energy.
All this shows we lack a lot of reliable data , in embodied energy, especially for heating ventilation and related equipment. For infrared panels /gasboiler/thermostats/smart meters/etc. separate data still to be found , and suggestions for sources welcome.
Note: the option of No installation:
Mind that there is also an option to avoid installations…. This has ben introduced and practiced by the architects office Baumschleger-Eberle in Austria, first in a few offices, and recently in a housing project . I wrote about the offices here: 
Table with some rough estimations as reference:
 Beyond 0-energy: paradigm shift in assessment, in Buildings, special issue low Carbon Building design, Rovers R., 2015 Volume 5 page 1-13 , http://www.mdpi.com/2075-5309/5/1/1
 Environmental impact evaluation of energy saving and energy generation: Case study for two Dutch dwelling types, Michiel Ritzen, T.Haagen, Ronald Rovers, Chris Geurts, in: Building and Environment 108 ·July 2016, DOI: 10.1016/j.buildenv.2016.07.020
 More_Connect: tool and explanation http://www.more-connect.eu/wp-content/uploads/2019/07/MORE-CONNECT_WP3_D3.2-Tool-on-energy-costs-and-PV_explanation-audio-presentation.zip
 Spatial and Behavioral Thermal Adaptation in Net Zero Energy Buildings: An Exploratory Investigation , Shady Attia , Sustainability Spetember 2020, http://dx.doi.org/10.3390/su12197961
 The significance of embodied energy in certified passive houses, Robert Crawford et all, International Journal of Civil, Environmental, Structural, Construction and Architectural Engineering Vol:7, No:6, 2013
and: IEA EBC Annex 57 Embodied Energy: All reports to be found here: http://www.iea-ebc.org/projects/project?AnnexID=57 The Berlin office is in report Subtask 4 pag 324 case studies.
 Detailed Assessment of Embodied Carbon of HVAC Systems for a New Office Building Based on BIM Christina Kiamili 1 , Alexander Hollberg, and Guillaume Habert, Sustainability, April 2020, https://www.mdpi.com/2071-1050/12/8/3372
Energy payback time (EPBT) and energy return on energy invested (EROI) of solar photovoltaic systems: A systematic review and meta-analysis . K.P. Bhandari et al. / Renewable and Sustainable Energy Reviews 47 (2015) 133–141
Small scale CHP (english): https://db.ecoinvent.org/reports/20_SmallCombinedHeatPower.pdf?area=463ee7e58cbf8
Ventilation systems (residential, german): https://db.ecoinvent.org/reports/25_Komfortlueftung.pdf?area=463ee7e58cbf8
KBOB recommendation 2009/1:2016 (english/italian, a german/french version is available too): https://www.kbob.admin.ch/dam/kbob/it/dokumente/Publikationen/Nachhaltiges%20Bauen/Archiv_2015-2019/2009_1-2016%20Oekobilanzdaten%20im%20Baubereich.pdf.download.pdf/2009_1-2016%20Oekobilanzdaten%20im%20Baubereich.pdf
EE calculation tool: EC3: https://buildingtransparency.org/auth/login
– The embodied CO2-e of sustainable energy technologies used in buildings: A review article
Stephen Finnegan ,Craig Jones, Steve Sharples in Energy and Buildings, 27 September 2018 , https://doi.org/10.1016/j.enbuild.2018.09.037
– The significance of embodied energy in certified passive houses, Robert Crawford et all, International Journal of Civil, Environmental, Structural, Construction and Architectural Engineering Vol:7, No:6, 2013
14 The relationship between operational energy demand and embodied energy in Dutch residential buildings , A. Koezjakov , D. Urge-Vorsatz, W. Crijns-Graus, M. van den Broek , Energy and Buildings january 2018, https://doi.org/10.1016/j.enbuild.2018.01.036
15 Embodied GHG emissions of buildings – The hidden challenge for effective
climate change mitigation , Martin Röck et all, Applied Energy, november 2019 https://doi.org/10.1016/j.apenergy.2019.114107
 Baumschleger-Eberle: houses without heating: https://archello.com/project/2226-graf-multiple-dwelling and